Proceedings of the 28th Annual Hawaii International Conference on System Sciences - 1995 Group Strategic Decision Making: Asynchronous GSS Using Structured Conflict and Consensus Approaches Jerry Fjermestad School of Industrial Management New Jersey Institute of Technology S. Roxanne Hiltz Murray Turoff Newark. New Jersey 07102 jeay@eies njkedu Robert M. Czech BMC Communications P.O. Box 7.843 Clifton, New Jew.9 07015 bmc.bi@hwtstmp.orS Computer k lnfommtion Sciences New Jen9 Institute of Technology Newar!c New Jersey 07102 roxann@eics.njit.cdu mumy@& njit.edu ABSTRACT This longitudinal experiment investigated the efficiency effectiveness and group member perceptions of dialectical inquiry (DI) and constructive consensus (CC) approaches to strategic decision making in an asynchronous (distributed) Computer-Mediated-Communications (CMC) environment. There were no dQ,%erences between DI and CC groups in terms of decision effectiveness. DI groups required signljicantly more asynchronous meeting time to complete their recommendations and had higher perceived depth of evaluation than CC groups. CC groups reported greater decision acceptance and willingness to work together again than DI groups. Relatively few experiential eflects were observed. The results are discussed in terms of their implications for Group Support Systems research and design. INTRODUCTION Researchon face-to-face (FtF) groups has indicated that groups that try to reach consensuson a decision without following any specific procedures, often suffer “processlosses” such as unequal participation, failure to generate and explore a number of options beforereaching a decision, and a lack of critical examination of ideas. In FtF groups procedures such as brainstorming, Nominal Group Technique, and structuredconflict procedures such as Dialectical Inquiry (DI) can decrease process losses and improve the outcomesof group decision making [29, 3 1, 321. In DI, the group as a whole is divided in half, with one subgroup charged with developing a plan or proposal, and the other sub-group charged with coming up with criticisms and a counter-proposal or counter-plan. Previous studies in the field of organizational strategic decision making have demonstratedthat DI and similar structuredconflict approaches, such as “devil’s advocate” (DA) procedures can improve the quality of decisions [22, 24, 3 1, 32, 331. To what extent do the findings about FtF groups apply to groups supported by computer based Group Cameron Ford Rosalie Ocker Frederick Ferront Malcolm Worrell Kenneth Johnson Department of Psychology Upsala College East Orange, New Jersey 0701.5 Graduate School of Management Rutgets Univmiiy Newark. New hsey 07102 Support Systems (GSS), both in the same-time/same place “decision room” situation, and in the distributed or asynchronous (different time/different place) condition? A study by Tung and Heminger [40] utilized synchronous GSS technologies in a single experimental session, and found no differences among DA, DI, and consensus proceduresin terms of group performance or subjective perceptions. There have been no studies of distributed (asynchronous) GSS’s that compare conflict and consensus approaches. One reason why Tung and Heminger may have found no differences for a synchronous GSS is that, as Churchman [4] Manson and Mitroff [23] and Mitroff et al [26] suggest, structured decisionapproaches, especially DI, take time and practice to learn. Thus, the results reported by Tung and Heminger are not surprising; their groups had to learn the GSS, the decision approach,and organize themselves,all in a single short experimental time period. The purpose of this study was to compare the effectivenessof a structured conflict approach (DI) with a structured consensus approach (which we called “constructive consensus” or CC), for strategic decision making tasks, for groups using a distributed, asynchronous GSS embedded in a computer conferencing system. We gave the groups a two-hour training with a practice FtF task first, to make sure that they would have time to leam and adapt to the GSS, the decision approaches, the type of task, and the group itself. Each group then had two different strategic decision tasks, each lasting for two full weeks of communication and decision making. A variable of secondary interest was the effect of “experience;” Chidambaram[2,3] had found that in a Decision room situation, groups improved significantly over time. Would this also be true in the much more extended Distributed GSS situation; eg, after extensive initial training, would there be any differences betweenan initial two-week task and a secondtwo week task? 222 1060-3425/95$4,00@1995IEEE Proceedings of the 28th Hawaii International Conference on System Sciences (HICSS '95) 1060-3425/95 $10.00 © 1995 IEEE